Arch Environ Contam Toxicol DOI 10.1007/s00244-014-0107-6

Time Trends in Serum Organochlorine Pesticides and Polychlorinated Biphenyls in the General Population of Biscay, Spain Miren Begon˜a Zubero • Juan J. Aurrekoetxea • Mario Murcia • Jesus M. Ibarluzea Fernando Gon˜i • Cristina Jime´nez • Ferran Ballester



Received: 27 February 2014 / Accepted: 20 November 2014 Ó Springer Science+Business Media New York 2015

Abstract Despite the measures adopted, levels of organochlorine compounds (OCs) are still being detected in the human body. This study aimed to explore factors related to changes in the concentrations of polychlorinated biphenyls (PCBs) and pesticide OCs in blood samples obtained from a general population cohort. Two cross-sectional samples were taken from 162 adults (2–75 years of age), with a gap of 2 years, from four areas in Biscay (Spain). More than 75 % had quantifiable levels of hexachlorobenzene (HCB), beta-hexachlorocyclohexane (b-HCH), dichlorodiphenyldichloroethylene, and PCBs 138 153 and 180. During this time, significant changes were observed: PCB 180 and HCB levels increased, and PCB 138 and b-HCH levels decreased. Regarding age, this study shows a decrease suggesting a cohort effect. The period was not related to the decrease of levels in all age intervals, but a statistically significant increase of PCBs in older people was found. High body mass index was associated with lower PCB 180 levels and

Electronic supplementary material The online version of this article (doi:10.1007/s00244-014-0107-6) contains supplementary material, which is available to authorized users. M. B. Zubero (&) Basque Foundation for Health Innovation and Research (Bioef), Sondika, Biscay, Spain e-mail: [email protected] M. B. Zubero  J. J. Aurrekoetxea Department of Preventive Medicine and Public Health, University of the Basque Country, Leioa, Biscay, Spain J. J. Aurrekoetxea  J. M. Ibarluzea Public Health Department, Basque Government, San Sebastia´n, Spain

greater HCH levels. Inversely, greater levels of HCB and bHCH were in those who had lost weight before the study. Levels of HCB and b-HCH were also greater in women who had had children, although they were lower in those who breastfed. Levels of these same OCs were greater in fish consumers, whereas those of PCBs 138 and 153 were greater in those who consumed local produce; all of these trends were close to significance. Efforts should continue to decrease exposure to these pollutants and to assess their influence on general population.

Abbreviations BMI Body mass index EPIC European prospective investigation into cancer and nutrition HCB Hexachlorobenzene HCH Hexachlorocyclohexane IARC International Agency for Research on Cancer LOQ Limit of quantification OCs Organochlorine compounds

M. Murcia  J. M. Ibarluzea  F. Gon˜i  F. Ballester Spanish Consortium for Research on Epidemiology and Public Health, Valencia, Spain M. Murcia  F. Ballester Center for Public Health Research, Valencia, Spain F. Gon˜i  C. Jime´nez Public Health Laboratory, Basque Government, San Sebastia´n, Spain F. Ballester Department of Nursing, University of Valencia, Valencia, Spain

J. J. Aurrekoetxea  J. M. Ibarluzea  F. Gon˜i Biodonostia Health Research Institute, San Sebastia´n, Spain

123

Arch Environ Contam Toxicol

PCBs p,p0 -DDE p,p0 -DDT REML a-HCH b-HCH c-HCH

Polychlorinated biphenyls Dichlorodiphenyldichloroethylene p,p0 -Dichlorodiphenyltrichloroethane Restricted maximum likelihood Alfa-hexachlorocyclohexane Beta-hexachlorocyclohexane Gamma-hexachlorocyclohexane

Organochlorine pesticides (OPs) are a group of chemical compounds that are effective against a wide variety of insects and other pests. The drawback of using them is their toxicity and high persistence. Among them, p,p0 -dichlorodiphenyltrichloroethane (p,p0 -DDT) and its main metabolites— dichlorodiphenyldichloroethylene (p,p0 -DDE), beta-hexachlorocyclohexane (b-HCH), and hexachlorobenzene (HCB)—are the most commonly found pesticides in the environment (WHO 2003). In the past, p,p0 -DDT was widely used as an insecticide in agriculture as well as in public health programmes against the malaria vector. Once absorbed, p,p0 -DDT is metabolized to p,p0 -DDE, a more stable and persistent form, in the body. As a consequence, this compound is considered a good marker of chronic exposure (Jaga and Dharmani 2003). Although prohibited in Spain in the seventies, p,p0 -DDT is still being used in some countries (Rogan and Chen 2005). As for HCH, although it has several isomers, only lindane (gamma-hexachlorocyclohexane) has insecticidal properties. In the production process of lindane, various HCH isomers are produced, and b-HCH is the most persistent. HCB was used as a fungicide, but today it is an industrial byproduct (WHO 2003). Last, heptachlor epoxide and endosulfan were used as an insecticide and acaricide, respectively. Endosulfan became a highly controversial agrichemical due to its acute toxicity and role as an endocrine disruptor (WHO 2003). Another group of organochlorine compounds (OCs) is polychlorinated biphenyls (PCBs). They are not naturally produced, and they have been used in many industrial applications including as dielectric fluids in capacitors and transformers and additives in pesticides, paints and plastifying agents, as well as pesticides. These compounds are also lipophilic and tend to bioaccumulate. There are [200 types of PCBs, those most common being 118 and, above all 138, 153 and 180. Notably, the International Agency for Research on Cancer (IARC) recently upgraded PCBs to group 1 as human carcinogens (IARC 2014). It is believed that the exposure of the general population to these compounds is decreasing (Hagmar et al. 2006; Porta et al. 2012; Petrik et al. 2006; Sjo¨din et al. 2004). Pesticides are used in agriculture, and the source of PCBs is industry, but the source of current exposure to both groups of substances is through food. The use of OPs was

123

restricted heavily in the 1970s and that of PCBs in 1980. Both are currently banned in Spain. Diet, through the fat of fish and meat as well as other fats, is the most common source of exposure of OC pesticides or PCBs in the general population (WHO 2003; Domingo and Bocio 2007). The load of these compounds in the body remains a concern across the world due to their potential adverse effects on health and the environment. In the Basque Country, the intake of pesticides through the diet was found to be very low, with values of \7 % of the tolerable daily intake (Jalo´n et al. 1997). In Spain, despite most OCs having been banned in the 1970s and 1980s, relatively recent studies have detected residues of OCs in foodstuffs as well as in the general population (Agudo et al. 2009; Carren˜o et al. 2007; Jakszyn et al. 2009; Luzardo et al. 2006; WHO 2003). Among the publications available, two studies are of particular interest: one analyzing the concentrations of OPs in serum in a representative sample of the general population in the Canary Islands (Zumbado et al. 2005) and the other assessing OPs and PCBs in serum in a representative sample of the population in Catalonia (Porta et al. 2010). In both studies, DDE was present in[80 % of the samples. In addition, it has been found that there is considerable prenatal exposure to these substances (Vizcaino et al. 2011). Given their high persistence and liposolubility, it is possible to study the presence of these compounds in adipose tissue, blood serum, and breast milk (LaKind et al. 2001). The advantage of serum is that it can be collected from any individual, although the amount of fat is lower than in other biological matrices. In relation to time trends, Petrik et al. (2006) observed a 30 % decrease in the levels of PCBs and OPs comparing two groups whose samples were taken in 1998 and 2002 in Slovakia et al. (2012) found 34–56 % decreases in OCs between 2002 and 2006 in Barcelona (Spain). Although these studies were performed in general-population samples, they were cross-sectional study designs with great interindividual variability, in which various confounding factors could influence the results. The cohort studies, however, have another limitation for investigating trends in OC levels in that, during the study period, the age of the participants also increases. With adjustments, considering these two time-dependent factors, age can be an indicator of the birth cohort and of the timescale for changes in OCs. Older cohorts have had more exposure, and the period is related to the balance between current exposure and metabolic clearance of the compound. As well as these two time-dependent variables, exposure is related to geographical area and sex. These four variables are clearly associated with the observed variability (Porta et al. 2008). In February 2006 and February 2008, the levels of dioxins, furans, and heavy metals were quantified in blood

Arch Environ Contam Toxicol

samples of the general population in Biscay (Basque Country) to evaluate their exposition from an urban waste treatment plant (Zubero et al. 2010, 2011). The aim of this study was to explore time-dependent factors, age, and, especially, changes in concentrations of PCBs and pesticides, and OCs quantified for reasons of opportunity, in a cohort composed of individuals from whom a second blood sample was obtained.

Methods Population and Sample Systematic random sampling was used, stratifying by age and sex, from the municipal census provided by the local councils of four areas in Biscay, two towns and two districts in the city of Bilbao, within a 60-km radius. Two of the four areas selected were located close to a waste incinerator (the town of Alonsotegi and the district of Rekalde in Bilbao), whereas the other two were further away (the district of Santutxu and the town of Balmaseda). A total of 322 and 326 volunteers were recruited in 2006 and 2008, respectively (Zubero et al. 2011). In relation to the origin of the samples, in 2006, 36 % of the participants came from the census in the areas close to the plant, and the rest were volunteers from the health center. In remote areas, 25 % were recruited from the census, and the rest were volunteers. In 2008, in nearby areas, 71 % were previous participants, whereas in the remote areas this percentage was 68 %. This analysis was based on the 162 individuals who completed both surveys. The two sampling were performed at the same time of the year (from January to April). All participants signed an informed consent form before participation in the study. The study was approved by the Clinical Research Ethics Committee of the local referral hospital. Sampling and Laboratory Analysis A blood sample was taken from each participant. They were not required to fast before the blood collection, but they were recommended to avoid eating fats the night before. Samples were taken, treated, stored, and transported in accordance with the conditions established by the reference laboratories (Patterson et al. 1991). Analyses were performed in the Public Health Laboratory of Gipuzkoa according to a previously described method (Gon˜i et al. 2007). Extraction was performed in solid phase of 500 ll of serum on C18 extraction discs arranged on 96-well plates, followed by purification of the extract using liquid–solid adsorption in silica/sulphuric acid columns, and then concentration to 50 lL in cyclohexane. Quantification was

performed using an HP 5980 gas chromatograph (Agilent, Palo Alto, California, USA) fitted with an electron capture detector, an automatic injector, and a DB-XLB capillary column (30 m 9 0.25 mm i.d. 9 0.25 lm stationary phase). Confirmation was performed using an Agilent 6890 gas chromatograph with automatic 2injector, DB-5MS capillary column (50 m 9 0.25 mm i.d. 9 0.25 lm stationary phase), and a quadruple HP 5973 mass spectrometer in selected ion monitoring mode with electron-impact ionization and two ions per component. PCBs 46 and 143 were used as internal standards in both chromatographs. In the starting validation, the method was checked respect to recovery and short-term precision (repeatability) by spiking a serum at two levels (n = 8/level). Recoveries of PCB congeners 28, 52, 101, 118, 138, 153, and 180 and OC pesticides HCB, b-HCH, p,p0 -DDE, and p,p0 -DDT varied from 60 to 120 %, and intraday relative SD from 1 to 11 %, with both depending on spiking level and compound. In each set of samples, two blanks were included as were a duplicate control serum sample and duplicate NIST Standard Reference Material 1589a (National Institute of Standards and Technology, Gaithersburg, Maryland, USA). The mean concentrations of PCB 118, PCB 180, and p,p0 -DDE were in agreement with the certified values, whereas those of PCB 138 and PCB 153 were 16 and 13 % lower than the reference values, respectively. The remaining compounds were lower than the LOD, which also agreed with the certified values. Interday relative SD was \15 % in all cases. The levels of the following eight OPs were analyzed: HCB, alpha-hexachlorocyclohexane (a-HCH), b-HCH, gamma-hexachlorocyclohexane (c-HCH), heptachlor 0 0 epoxide, endosulfan beta, p,p -DDE, and p,p -DDT. The levels of the most common PCBs (PCB 28, 52, 101, 118, 138, 153, and 180) were also measured. The limits of quantification (LOQs) at both surveys were 0.10 ng/ml for PCBs and HCH isomers and 0.20 ng/ml for the other pesticides. The final results were corrected for enzyme-based measurements of lipid concentration and were expressed in nanograms of compound per gram of lipid (ng/g lipid). To express the concentration of the chemical residue as a function of the lipid content of the samples, we applied the algorithm of Philips et al. (1989). The adjustment is recommended because lipid changes in serum can be detected over time and give a more stable estimate (Koppen et al. 2002), especially considering the need to allow for the contribution of the weight of the lipids in the calculation (Bernert et al. 2007). Study Variables As time-dependent variables, we considered age and study period itself. We recorded sex and geographical area; body mass index (BMI) and categorized participants into three

123

Arch Environ Contam Toxicol

groups; any weight gain or loss C5 kg in the last 5 years before the survey; number of children; and, in the case of women, whether or not they were breastfeeding. In addition, socioeconomic status was defined using the Spanish adaptation of the British Registrar General classification system (Domingo-Salvany et al. 2000) grouping the social classes into three categories: high (classes I and II), medium (class III), and low (classes IV and V). Education level was also grouped into three categories: primary or less, secondary, and university. Information was also collected on whether individuals often consumed locally grown produce (yes or no) and whether they consumed saltwater fish grouped into three categories: 0–1, 2–4, and [4 times per week.

The presence of interactions with the survey/time point variable was explored to assess whether any study variables interfered with the changes in OCs levels over time. The approximate normality of residuals was confirmed graphically. We also explored whether there was a linear relationship between the variable of age and the concentration of the compounds analyzed. To avoid individual observations having an excessive influence on the results, the cases with intraindividual Pearson residuals [4 in absolute value were removed from the final models. The level of significance was set at a = 0.05. Data were analyzed using the SPSS version 17.0 for Windows (SPSS, Chicago, IL, USA) statistical package and R version 2.15.3 [R Foundation for Statistical Computing, Vienna, Austria (http://www.R-pro ject.org)].

Statistical Analysis Before correcting for lipids, samples that were lower that the LOQ were replaced with the LOQ divided by the square root of 2 (CDC 2005). Statistical analysis included OCs with quantification frequency [75 %. Logarithmic transformation of the lipid-corrected data brought it closer to a normal distribution and decreased the variance. For hypothesis testing, McNemar test was used for qualitative variables. Geometric means and 95 % confidence intervals (CIs) were calculated, and Student t test for paired samples was used for comparing the means of the biological indicators between 2006 and 2008 in the each year’s analysis. Student t test for independent samples and analysis of variance were employed for comparing means of variables with two or more categories, respectively, using the trend test for ordinal variables in the each year’s analysis. Determinants of the OCs were explored taking into account the intraindividual correlation between surveys using mixed-effects linear models (Pinheiro and Bates 2000). The concentration of each compound (expressed in nanograms per milliliter) was logarithmically transformed and adjusted using a specific model. Starting from a full model, variables were excluded using a backward selection process using a p value [0.10 in the likelihood ratio test as the exclusion criterion. The lipid variable was considered as an independent term of the models (Schisterman et al. 2005). In all of the models, regardless of their statistical significance, we adjusted for the following variables: lipids (logarithmically transformed), time point/survey, age, geographical area, and sex. Final models were estimated by restricted maximum likelihood. The percentage of the unexplained variance that was attributable to the individual (intraclass correlation coefficient) was estimated as the quotient between the estimate of the variance of the intercept random effects and the sum of the variance of the intercept random effects plus the variance of the within-group errors, q = r2between /(r2between ? r2within) (Pinheiro and Bates 2000).

123

Results Table 1 lists the characteristics of the 162 participants of the 2006 and 2008 surveys. Table 2 lists the number of samples with values greater than the LOQs. We did not detect PCBs 52 or 101, endosulfan beta, heptachlor epoxide, or a-HCH in any of the surveys. Nevertheless, all individuals had a quantifiable amount of at least one OC. In the second survey, we found a greater proportion of cases greater than the LOQ of PCBs 28 (p = 0.006) and 118 (p \ 0.001) and a lower proportion of detectable cases of p,p0 -DDT (0.007) and ß-HCH (p = 0.015). Results for PCBs 28, 52, 101, and 118, as well as heptachlor epoxide, c-HCH, and p,p0 -DDT, were quantifiable in \75 % of the samples in both years; hence, these were excluded from the subsequent statistical analysis. Analyzing the mean levels of the three PCBs and three OC pesticides most commonly found ([75 % greater than the LOQ), and the changes therein between the measurements at the two time points, we observed that p,p0 -DDE had the highest mean concentrations, followed by HCB and PCBs 153 and 180, in both years (Table 3). During the 2-year period, we observed a significant increase in the mean levels of PCB 180 and HCB and a significant mean decrease of PCB 138 and b-HCH. The Supplementary Table shows the relationship between the socioeconomic variables and the mean values of OCs for the two survey years. Age was significantly positively associated with the concentration of all of the OCs (p \ 0.001 for all of the OCs). We found statistically significant differences between the different areas with a nonhomogenous geographical pattern. Men had significantly greater levels of PCB 180, whereas women had significantly greater levels of HCB and ß-HCH. Women who had had any children showed significantly greater mean levels of OCs than women who had not, and there

Arch Environ Contam Toxicol Table 1 Description of the study sample

Table 1 continued

Variable

N

(%)

All participants

162

100

Variable

Origin of the sample 2006 Census volunteers

50

30.9

Neighbourhood volunteers

70

43.2

Health centres

42

25.9

2008 Previous volunteers

113

69.8

Neighbourhood volunteers

18

11.1

Health centres

31

19.1

Zone Alonsotegi

44

27.2

Rekalde

30

18.5

Santutxu

37

22.8

Balmaseda

51

31.5

Woman

102

63.0

Man

60

37.0

47.0

12.6

20–44

73

45.1

45–69

89

54.9

26.4

4.9

\25

71

43.8

25–29

Sex

Age (year) (mean SD)a Age range in years (mean SD)a

BMI (mean SD) BMI group

63

38.9

C30 Weight changeb

28

17.3

Weight gain

27

16.7

No change

126

87.7

Weight loss

9

5.6

Level of education University or other tertiary

44

27.2

Secondary

54

33.3

Primary or less

64

39.5

I–II (highest)

37

12.8

III

32

19.8

IV–V (lowest)

93

57.4

0

92

56.8

1–2 C3

54 16

33.3 9.9

No

15

21.4

Yes

55

78.6

27

16.7

Social class

No. of children

Had children and breastfed

Consumption of fish (portions/week) 0–1

N

(%)

2–4

121

74.7

C5

14

8.6

71

43.8

91

56.2

Consumption of locally grown produce No Yes a

Mean (SD) age at the first survey

b

Body weight change in the 5 years (C5 kg) before the first survey

was no statistically significant difference between those who had and had not breastfed their children. With respect to BMI, there was a significant gradient, with greater mean levels of HCB, b-HCH, and p,p0 -DDE, in obese and overweight people. Body weight change before the first survey was related to a significant trend with greater levels of the three PCBs (138, 153, and 180) in those who had lost weight. Individuals from the lowest social class and with the lowest level of education had significantly greater levels of all of the OCs, except for p,p0 -DDE, wherein no differences were observed. The consumption of local produce was not significantly associated with the concentration of OCs, but the frequency of fish consumption was positively associated with the concentration of all OCs except for PCB 180. Figure 1 shows the distribution of OCs in 2006 and 2008 (period with an increase in age of 2 years) by age (year of birth or cohort). All OCs showed a clear decrease of levels over the date of birth. Only ß-HCH showed decrease of levels across the sampling period in all cohorts. Among older people levels of PCBs and HCH increased. After adjustment (Table 4), between 2006 and 2008 (sampling time) there was a significant decrease in the levels of PCBs 138 and 153, ß-HCH, and p,p0 -DDE but not of PCB 180 or HCB. The levels of all of the OCs significantly increased with age (p \ 0.01). By area of residence, we observed a high geographical variability with significant differences for PCB 180, HCB and p,p0 -DDE—the pattern being consistent for PCB 180 and HCB—with greater levels in the most rural area, Balmaseda. As shown before adjustment, there were significant greater levels of PCB 180 in men than in women, but HCB and ß-HCH levels were significantly greater in women. Greater BMI levels were significantly related to low levels of PCB 180 and to high levels of HCB. A significant increase in the levels of HCB and ß-HCH was associated with having lost weight before the first survey. Similarly, the levels of these two pesticides were significantly greater in women who had children, but significantly lower in those who had breastfed, in relation to those women who did not have

123

Arch Environ Contam Toxicol Table 2 Blood serum samples with values greater than the LOQ

Compounds

p value from exact McNemar test comparing the detection of compounds in 2006 and 2008; p values were two-sided

2006 n

pa

2008 (%)

n

(%)

PCB 28

0.10

1

0.6

PCB 52

0.10

0

0

PCB 101

0.10

0

0

0

0

PCB 118

0.10

49

30.2

81

50.0

PCB 138

0.10

161

PCB 153

0.10

162

PCB 180

0.10

161

0.20

0

HCB

0.20

135

a-HCH

0.10

0

b-HCH c-HCH

0.10 0.10

142 2

p-p0 -DDT

0.20

28

17.3

15

9.3

0.007

p-p0 -DDE

0.20

160

98.8

158

97.5

0.500

Hepta chlorepoxide

a

LOQ (ng/g lipid)

99.4 100 99.4 0 83.3 0 87.7 1.2

10 0

6.2

0.004

0

– – \0.001

160

98.8

1

160

98.8



162

100



0

0



145

89.5

0

0

129 6

79.6 3.7

0.013 1 0.015 0.219

Table 3 Geometric means and 95 % CIs for OC concentrations measured in 2006 and 2008 Compounds

2006 GM

% cha

2008 95 % CI

GM

pb

95 % CI

PCB 138

68.0

62.2–74.3

64.8

58.7–71.6

–4.6

0.030

PCB 153

93.8

85.2–103.3

92.4

83.2–102.7

–1.5

0.500

PCB 180

85.6

78.3–93.6

92.4

84.0–101.7

7.9

0.004

HCB

99.3

85.7–115.1

107.2

92.7–124.1

8.0

0.028

b-HCH p-p0 -DDE

46.2

40.6–52.5

36.2

31.6–41.4

–21.7

\0.001

191.8

165.8–221.9

187.4

164.2–213.7

–2.3

0.508

Compounds were quantified in[75 % of individuals in both surveys. Percentage change and comparison of means were applied to the logarithm of the concentrations. Concentrations are expressed as nanograms per gram of lipid a

%ch = 100 * ((GM OC2008/GM OC2006)-1): percentage change of OCs are associated with the period of sampling

b

Student t test for comparing the means of paired samples

children and who had not breast-feed them, respectively. In addition, there were significant increases in the levels of these two pesticides among those who consumed saltwater fish. We did not observe any significant associations between the levels of OCs and consumption of local produce, social class, or level of education. Intraclass correlation coefficients ranged between 0.73 and 0.83 for PCBs, HCB, and ß-HCH, and 0.95 for p,p0 -DDE, thus indicating the increased persistence of these compounds in the organism. Table 5 lists the interactions observed between the sampling time when blood samples were taken for analysis and the variables entered in the regression models of Table 4. It indicates whether changes in various subgroups were significantly different. No significant interactions were observed in the case of ß-HCH. We found a significant interaction between sampling time and age, were older

123

people had increased levels of PCBs but decreased levels of p,p0 -DDE. In addition, we observed that changes in OC level varied among geographical areas with a trend toward greater decreased levels in Alonsotegi and Rekalde and less decreased levels in Santutxu and Balmaseda, where PCB 180 levels actually increased significantly. Other variables, such as parity, breastfeeding, and BMI were not associated with changes in OC levels. In contrast, there were trends toward levels of HCB decreasing in those who had lost weight and increasing significantly in those who had gained weight.

Discussion All individuals had detectable levels of any of the analyzed OCs in their blood. Moreover, PCBs 138, 153, and 180 and

Arch Environ Contam Toxicol

Fig. 1 OC distribution (ng/g lipid) by cohort (birth year [age]) and period (sampling year [age]) in 162 individuals (paired samples)

pesticides HCB, ß-HCH, and p,p0 -DDE were detectable in [75 % of individuals. The presence of samples greater than the LOQ in blood and mean levels of PCB 180 and HCB increased during the study period. Only the levels of ß-HCH and PCB 138 decreased significantly. For all of the other OCs, there were no significant changes. The fact that the use of these substances is prohibited in our setting is not sufficient for them to totally disappear from our bodies or even that the balance between intake and clearance is such to favour to their elimination from the body. Porta et al. (2012), Petrik et al. (2006), Sjo¨din et al. (2004) and Hagmar et al. (2006) observed more marked decreases in OCs than we found in this study. Nevertheless, this study only covered a period of 2 years. It included repeated measurements in the same individuals, and hence the results show the balance between OC intake and metabolism. Hagmar et al. (2006) also observed decreases in levels of PCB 153 by 34 %, p,p0 -DDE by 55 %, and HCB by 53 % in a sample of 39 individuals over 10 years. In contrast, Tee et al. (2003) did not observe a decrease of total PCBs among 78 not freshwater fish consumers from 1980 to 1994 but only freshwater fish consumers with a decrease of fish consumption. Hovinga et al. (1992) did not observe decrease of PCB levels among 95 non-freshwater fish consumers

from 1982 to 1989 with only a small decrease among those who did consume freshwater fish. In our study, there was a marked association between age and all of the OCs suggesting that these compounds are increasing in the body (age effect). However, the design of this study, i.e., with repeated measures on the same individuals, shows that this increase does not coincide with the actual time, which suggests more a cohort effect as Porta et al. (2008) stated. Nøst et al. (2013) observed throughout 5 surveys of a cohort of 53 men a decrease in the levels of PCB 153 in relation to age, controlling for birth date, using an age–cohort period graphic analysis. The association between age and OCs has been widely reported in the literature (Porta et al. 2008; Jakszyn et al. 2009). Furthermore, in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort in Spain, the levels found in samples collected between 1992 and 1996 for PCBs (Agudo et al. 2009) and pesticides (Jakszyn et al. 2009) were greater, especially in the Gipuzkoa subset, than in this study for all of the pesticides and PCBs except for PCB 180. Gipuzkoa region is close to Biscay (both are in the Basque Country, Spain) with similar social and industrial structure as well as diet. In contrast, the EPIC study used blood samples of people age between 35 and 65 years old (birth year from 1927 to 1961), thus

123

123

–9.4

5.1

22.4

16.9

Rekalde

Santutxu

Balmaseda

3.1–4.2

8.8

3.6

Male

\0.001

\0.001

0.091

0.060

0.003

\0.001

3.5

26.0



24.0

8.4

-4.9



3.0



77.3

-32.5

Yes

– 19.5 47.3

0–1

2–4

C5

Fish consumption (portions/ week)



2.1–112.6

-5.2 to 50.7

-51.5 to -6.1 0.093

51.8

11.5



-32.0



57.8 0.017

70.0

8.8



4.9

88.7

0.001

\0.001

0.031

\0.001

– 31.0–171.7

54.9–226.9

-23.5 to 25.2

7.9–80.5

-13.1 to 31.5

3.7–5.5



No

Breastfed

Yes

No

Had children

-2.1 125.0

Gain

39.5

6.9



4.6

-27.6

-13.6

9.0

-0.3



-29.0



113.3

%ch

-43.4

\0.001

\0.001

0.736

0.021

p

– -57.1 to -25.4

12.7–77.8

-6.4 to 51.7

-28.0 to 19.9

-5.1 to 7.7

4.4–74.4

95 % CI

ß-HCH



41.5

19.2

-7.1



1.1



34.9

%ch

Loss

-37.4 to -9.8

0.003

\0.001

\0.001

0.019

0.271

\0.001

p



-24.8

-29.5 to -5.0

2.9–4.0

10.1–44.3

4.9–46.6

-9.3 to 29.4

-21.3 to 15.0

-2.2 to 8.4

44.6–117.5

95 % CI

HCB

None

Weight change

C30

-18.2

3.3–4.5

-2.1 to 30.8

-0.7 to 44.7

0.8–53.9

-19.5 to 24.2

-10.7 to -2.3

43.5–109.9



3.9

13.2



19.9

24.5

0.0



–6.6



73.6

%ch

25–29

\0.001

0.220

0.144

\0.001

p

PCB 180

\25

BMI (kg/m2)

Age (year)

-5.1 to 24.7



-2.1 to 39.6

0.3–49.4

-14.3 to 29.0

-13.3 to -5.4

50.2–116.5

Female

Sex



Alonsotegi

Area

-

2008

80.3

95 % CI

%ch

pb

%cha

95 % CI

PCB 153

PCB 138

2006

Time point

Lipids (mg/dl)c

Variables

Table 4 Variables associated with PCBs and pesticides

7.3–114.8

–10.5 to 38.9

-50.2 to -7.0

11.4–123.5

19.0–142.9

-13.2 to 36.3

4.0–5.7

-43.7 to -6.9

-30.5 to 7.3

-13.4 to 37.2

-21.8 to 27.0

-33.5 to -24.1

64.0–177.3

95 % CI

0.054

0.014

0.009

0.012

\0.001

0.011

0.203

\0.001

\0.001

p

4.2

-5.0



19.2

63.6

17.6



-6.0



31.1

%ch

3.2–5.2

-25.2 to 20.7

-12.1 to 61.7

18.5–125.8

-17.4 to 67.4

-10.0 to -1.8

8.8–58.1

95 % CI

p-p0 -DDE

\0.001

0.669

0.026

0.006

0.005

p

Arch Environ Contam Toxicol

14.3

0.82

ICCd

-2.3 to 33.7 0.83

16.1

– -1.7 to 37.1 0.73

%ch

95 % CI

p

0.77

%ch

HCB 95 % CI

p

0.73

%ch

ß-HCH 95 % CI

p

0.95

%ch

95 % CI

p-p0 -DDE

p value from the likelihood ratio test

ICC = intraclass correlation coefficient: q ¼ r2between =ðr2between þ r2within Þ

Natural logarithm of lipid concentration

-9.9

-20.0

-6.4

-1.6

-10.3

1.0

-1.5

-17.7

20.3

20.3

2.8

7.9

0.9

0.015 0.038

-31.6

2.5



14.4

40.7



95 % CI

0.043

p

2.3

-5.2

-18.2

-7.7

-0.4

%ch

-4.4

-12.2

-25.2

-14.0

-0.7

9.6

2.3

-10.5

-0.8

-0.1

95 % CI

p-p0 -DDE

c

b

\0.001

0.005

p

The coefficients represent the joint effect of time point and area to improve interpretability. The p value is that for interaction

p value from the likelihood ratio test

% ch = (exp(b)-1) 9 100; percentage change in OCs levels were associated with a change from the reference to current category or with an increment of one unit in continuous variables

The models included the same variables listed in Table 4, and the interactions with time point were additionally included if they had a p value from the likelihood ratio test \0.10

-11.5

10.3

8.9

-8.0

0.1

%ch

Loss

4.9

10.3

-2.3

0.5

p

20.1

-9.7

-7.0

-19.2

0.003 0.026

95 % CI



-2.7

1.2

-11.2

0.8

%ch

HCB

Gain

-0.7

4.5

1.5

0.2

p

PCB 180

None

Time point 9 weight change

-11.5 -14.1

-3.8

-7.7

Santutxu

Balmaseda

-22.7 -15.6

-16.6

-7.4

Alonsotegi

0.9 -13.5

0.5

0.2 0.071

\0.001

0.5

95 % CI

%ch

pb

%cha 95 % CI

PCB 153

PCB 138

Rekalde

Time point 9 areac

Time point 9 age (year)

Variables

a

p

%ch = (exp(b)-1) 9 100; percentage change in OCs were associated with a change from the reference to the current category or with an increment of one unit in continuous variables

0.076

p

PCB 180

Table 5 Variables associated with the change in OC concentrations between 2006 and 2008: Evaluation of the interaction with time point in 162 individuals (paired samples)

d

c

b

a

Mixed-effects linear regression model in paired samples



Yes

0.092

95 % CI

%ch

pb

%cha

95 % CI

PCB 153

PCB 138

No

Consumes local produce

Variables

Table 4 continued

Arch Environ Contam Toxicol

123

Arch Environ Contam Toxicol

increasing the comparability between both studies. This suggests that the association observed in these studies is not only due to an accumulative effect with age but are also due to a cohort effect. This study provides analysis not only of changes in OCs levels but also of interactions with the sampling time point. Older individuals had increased levels of PCBs but decreased levels of p,p0 -DDE. An explanation for this observation would be that the current exposure of PCBs were greater in older individuals. In contrast, the interaction was negative in the case of p,p0 -DDE, suggesting a faster decrease among elders, i.e., individuals with greater levels of these compounds. In our study, levels of PCBs were lower and those of pesticides were greater among women than among men as reported by other investigators (Agudo et al. 2009; Cerna´ et al. 2008; Charlier and Plomteux 2002; Costopoulou et al. 2006; Jakszyn et al. 2009; Jo¨nsson et al. 2005; Park et al. 2007; Petrik et al. 2006; Porta et al. 2010; Tee et al. 2003), although this is not a constant finding (Bates et al. 2004; Cruz et al. 2003; Lino and da Silveira 2006; Porta et al. 2012). In contrast, in this study, we did not find differences in the evolution of OCs as a function of sex. Another aspect to take into account to understand the distribution of OCs as geographical variability (Bachelet et al. 2011; Porta et al. 2012). Levels of OC pesticides in our study were lower than those found in other areas in Spain, namely, Catalonia (Porta et al. 2010), Andalusia (Carren˜o et al. 2007; Cerrillo et al. 2006), and the Canary Islands (Zumbado et al. 2005; Luzardo et al. 2006). In contrast, levels of PCBs were slightly greater than those in Catalonia (Porta et al. 2010) and much greater than those in the Canary Islands (Henrı´quez-Herna´ndez et al. 2011). Both of these observations may be explained by the fact that the Basque Country has a low level of agricultural activity but a considerable industrial sector. In our study, after adjustments, we observed significant differences between urban districts and towns \60 km apart with different patterns of changes in levels over time. That is, the balance between intake and metabolic clearance of OCs did not seem to be homogenous across the groups, even in closely located areas, suggesting a different pattern of exposure as a function of the area, given that the elimination should be similar, after adjustment for confounding variables. PCBs and OC pesticides are not related to emissions from waste incinerators. In this study most of the pollutants showed greater means in the areas far away from the plant. The consumption of saltwater fish was associated with greater levels of all of OCs studied except for PCB 180. In the Basque Country, there is high consumption of fish (81.21 g/day [Government of the Basque Country 2007]). Indeed, in the EPIC study, comparing 19 zones in 10 European countries, it was concluded that the consumption

123

of fish and fish products was highest in the Basque Country, this being mainly white fish (Welch et al. 2002). Kvalem et al. (2009) reported that the greatest source of PCBs (and dioxins) was oily and semi-oily fish. In a review of Spanish studies, Gasull et al. (2011) concluded that the consumption of animal products (fish, milk, and dairy products and meat) was associated with greater levels of OCs in the body. Greater consumption of fish (saltwater fish) has also been associated with greater levels of PCBs and p,p0 -DDE (Bachelet et al. 2011) as well as PCBs (Bra¨uner et al. 2011; Llop et al. 2010) and OPs (Bra¨uner et al. 2012). Petrik et al. (2006) found an association between decreased levels of PCBs and OPs and decreased intake of fish in their second study survey. Similarly, Hagmar et al. (2006) related decreases in levels of PCB 153, HCB, and p,p0 -DDE to changes in BMI and the consumption of Baltic Sea fish. Another Norwegian study reported an association between the consumption of fresh fish liver oil and seagull eggs in a sample of people with a diet based on seafood with greater levels of PCBs but not of p,p0 -DDE (Rylander et al. 2009). Similarly, the EPIC study (which was designed to assess the effect of diet on health) found a significant association between the concentration of PCBs and the consumption of fish (Agudo et al. 2009). In contrast, other research has not found an association between the levels of pesticides and diet (Jakszyn et al. 2009; Ibarluzea et al. 2011). In short, it seems that the consumption of fish is associated with increased levels of OCs, in particular PCBs, in the body. In contrast, our study did not show that eating fish changed the trend in OC levels over time. Fish consumption may be related to body burden rather than changes therein. No association was found among consumers of locally grown products and PCBs or pesticides. Bra¨uner et al. (2012) found an association between levels of OPs and the consumption of fruits and vegetables. In any case, the consumption of local products was not associated with an increase in OC levels over the study period, thus suggesting that it is not currently a significant source of these compounds. Another controversial issue in the literature regarding OCs has been the influence of body weight. This study found, after adjusting for other variables, greater levels of HCB in individuals with high BMI, whereas the opposite pattern was observed for PCB 180. Porta et al. (2012) did not find a clear pattern regarding the relationship between BMI and OCs, whereas Bra¨uner et al. (2012) observed greater levels of pesticides with greater BMI as we found in this study. In contrast, Bachelet et al. (2011) found a negative association between PCB concentrations and BMI in women as we found for PCB 180 in our study. Tee et al. (2003) observed also lower levels of total PCBs in individuals with high BMI.

Arch Environ Contam Toxicol

In any case, there was a consistent pattern with loss of body weight before the first survey with an increase of levels for HCB and ß-HCH. In contrast, we found a significant increase in levels of HCB from 2006 to 2008 in those who had previously gained weight. Porta et al. (2012) also noted lower levels of OCs in individuals who had gained weight in the previous 6 months, although these results were not adjusted. Bachelet et al. (2011) found (1) that in women, after adjustments, the greater the BMI, the greater the levels of p,p0 -DDE and the lower the levels of PCBs; and (2) there was a decrease in these compounds in people who had gained weight in the previous 10 years. Therefore, the role of BMI in determining OC levels, or even the direction of the association, is not clear (Moysich et al. 2002). There may be differences in the metabolism of OCs as a function of BMI. Wolff et al. (2005) suggested that greater quantity of fat in obese people may imply that metabolic clearance of OCs from the exposure earlier in life would be slower. Certainly, BMI and weight change do seem to modify the concentration of pollutants in blood (Carren˜o et al. 2007). Parity and breastfeeding were found to be associated only with levels of ß-HCH and HCB after adjustment with, an increment and a decrement, respectively. Porta et al. (2012) did not observe significant differences in the concentrations of OCs either of these variables. Similarly, the EPIC study did not find a significant trend between PCBs and maternal breastfeeding and even found lower levels in those who did not breastfeed (Agudo et al. 2009). Ibarluzea et al. (2011) and Llop et al. (2010) showed a decrease of levels of OCs in women who had breastfed during a long period of time. Social position, as measured by social class or by level of education, has been associated with levels of OCs found in biological samples (Ibarluzea et al. 2011; Porta et al. 2012). Nevertheless we did not observe such an association between social position and levels of OCs in our study after adjusting for age, time point, sex, and geographical area. The cohort nature of this study restricts the interindividual variability, which is characteristically high in crosssectional studies. In contrast, some aspects—such as the small sample size, the fact that subjects were volunteers, and the short time period between the two surveys (2 years)—could decrease the reliability of the study. Nevertheless, due to difficulties in the recruitment of and, especially, the maintenance of the willingness among the participants with repeated measurements, it would have been even harder to conduct this study with a longer period between surveys. To conclude, these results indicated that the general adult population of Biscay has been exposed to a greater or less extent to OCs and continues to be exposed to many of these compounds. In this population, the levels of these persistent pollutants are within the range observed by other

investigators, and we suspect that there is low-level diffuse exposure, probably through diet. We observed a downward trend in the levels of OCs in the general, the exception being PCBs and HCB in older people. The measures taken after the banning of these compounds or adoption of the Stockholm Convention seem to have had a beneficial effect, but there is ongoing exposure to OCs, the effects of which are still unknown. Further research may by necessary to elucidate how dietary habits or other factors could explain the differences in the evolution of OC levels as well as the effects of this exposure on health. Acknowledgments The authors express their gratitude to Zabalgarbi S. A, the company owning the waste incineration plant, and also to the Basque Health Service-Osakidetz, for providing access to their Health Care Centres and especially to all of the participants who kindly accepted to take part. This study was financed by Zabalgarbi S. A., the company owning the waste incineration plant, located in Bilbao, Biscay, Spain.

References Agudo A, Gon˜i F, Etxeandia A, Vives A, Milla´n E, Lo´pez R et al (2009) Polychlorinated biphenyls in Spanish adults: determinants of serum concentrations. Environ Res 109:620–628 Bachelet D, Truong T, Verner MA, Arveux P, Kerbrat P, Charlier C et al (2011) Determinants of serum concentrations of 1,1dichloro-2,2-bis(p-chlorophenyl)ethylene and polychlorinated biphenyls among French women in the CECILE study. Environ Res 111:861–870 Bates MN, Buckland SJ, Garrett N, Ellis H, Needham LL, Patterson DG Jr., et al (2004) Persistent organochlorines in the serum of the non-occupationally exposed New Zealand population. Chemosphere 54:1431–1443 Bernert JT, Turner WT, Patterson DG Jr., Needham LL (2007) Calculation of serum ‘‘total lipid’’ concentration for the adjustment of persistent organohalogen toxicant measurements in human samples. Chemosphere 68:824–831 Bra¨uner EV, Raaschou-Nielsen O, Gaudreau E, Leblanc A, Tjønneland A, Overvad K et al (2011) Predictors of polychlorinated biphenyl concentrations in adipose tissue in a general Danish population. Environ Sci Technol 45:679–685 Bra¨uner EV, Raaschou-Nielsen O, Gaudreau E, Leblanc A, Tjønneland A, Overvad K et al (2012) Predictors of adipose tissue concentrations of organochlorine pesticides in a general Danish population. J Expo Sci Environ Epidemiol 22:52–59 Carren˜o J, Rivas A, Granada A, Lopez-Espinosa JM, Mariscal M, Olea N et al (2007) Exposure of young men to organochlorine pesticides in Southern Spain. Environ Res 103:55–61 Centers for Disease Control and Prevention CDC (2005) Third national report on human exposure to environmental chemicals, Atlanta (GA). http://www.jhsph.edu/research/centers-and-insti tutes/center-for-excellence-in-environmental-health-tracking/ Third_Report.pdf. Accessed 20 Feb 2014 Cerna´ M, Maly´ M, Grabic R, Bata´riova´ A, Smı´d J, Benes B (2008) Serum concentrations of indicator PCB congeners in the Czech adult population. Chemosphere 72:1124–1131 Cerrillo I, Olea-Serrano MF, Ibarluzea JM, Exposito J, Torne P, Laguna J et al (2006) Environmental and lifestyle factors for

123

Arch Environ Contam Toxicol organochlorine exposure among women living in Southern Spain. Chemosphere 62:1917–1924 Charlier CJ, Plomteux GJ (2002) Determination of organochlorine pesticide residues in the blood of healthy individuals. Clin Chem Lab Med 40:361–364 Costopoulou D, Vassiliadou I, Papadopoulos A, Makropoulos V, Leondiadis L (2006) Levels of dioxins furans and PCBs in human serum and milk of people living in Greece. Chemosphere 65:1462–1469 Cruz S, Lino C, Silveira MI (2003) Evaluation of organochlorine pesticide residues in human serum from an urban and two rural populations in Portugal. Sci Total Environ 317:23–35 Domingo JL, Bocio A (2007) Levels of PCDD/PCDF and PCBs in edible marine species and human intake: a literature review. Environ Int 33:397–405 Domingo-Salvany A, Regidor E, Alonso J, Alvarez-Dardet C (2000) Proposal for a social class measure, Working Group of the Spanish Society of Epidemiology and the Spanish Society of Family and Community Medicine. Aten Primaria 25:350–363 Gasull M, Bosch de Basea M, Puigdome`nech E, Pumarega J, Porta M (2011) Empirical analyses of the influence of diet on human concentrations of persistent organic pollutants: a systematic review of all studies conducted in Spain. Environ Int 37:1226–1235 Gon˜i F, Lo´pez R, Etxeandia A, Milla´n E, Amiano P (2007) High throughput method for the determination of organochlorine pesticides and polychlorinated biphenyls in human serum. J Chromatogr, B: Anal Technol Biomed Life Sci 852:15–21 Hagmar L, Wallin E, Vessby B, Jo¨nsson BAG, Bergman A, Rylander L (2006) Intra-individual variations and time trends 1991–2001 in human serum levels of PCB DDE and hexachlorobenzene. Chemosphere 64:1507–1513 Henrı´quez-Herna´ndez LA, Luzardo OP, Almeida-Gonza´lez M, Alvarez-Leo´n EE, Serra-Majem L, Zumbado M et al (2011) Background levels of polychlorinated biphenyls in the population of the Canary Islands (Spain). Environ Res 111:10–16 Hovinga ME, Sowers M, Humphrey HE (1992) Historical changes in serum PCB and DDT levels in an environmentally-exposed cohort. Arch Environ Contam Toxicol 22:362–366 Ibarluzea J, Alvarez-Pedrerol M, Guxens M, Marina LS, Basterrechea M, Lertxundi A et al (2011) Sociodemographic reproductive and dietary predictors of organochlorine compounds levels in pregnant women in Spain. Chemosphere 82:114–120 International Agency for Research on Cancer (2014) Complete list of agents evaluated and their classification. http://monographs.iarc. fr/ENG/Classification/index.php. Accessed 20 Feb 2014 Jaga K, Dharmani C (2003) Global surveillance of DDT and DDE levels in human tissues. Int J Occup Environ Health 16:7–20 Jakszyn P, Gon˜i F, Etxeandia A, Vives A, Milla´n E, Lo´pez R et al (2009) Serum levels of organochlorine pesticides in healthy adults from five regions of Spain. Chemosphere 76:1518–1524 Jalo´n M, Urieta I, Macho ML, Azpiri M (1997) Vigilancia de la contaminacio´n Quı´mica de los Alimentos en la Comunidad Auto´noma del Paı´s Vasco: 1990-1995, Servicio central de publicaciones, Gobierno Vasco, Vitoria-Gasteiz. http://www. osakidetza.euskadi.net/r85-ckdrog02/es/contenidos/informacion/ sanidad_alimentaria/es_1247/seguridad.html. Accessed 20 Feb 2014 Joint WHO/Convention Task Force on the Health Aspects of Air Pollution (2003) Health risks of persistent organic pollutants from long range transboundary air pollution. World Health Organisation Regional Office for Europe, Copenhagen, Denmark. http://www.euro.who.int/__data/assets/pdf_file/0009/786 60/e78963.pdf. Accessed 20 Feb 2014 Jo¨nsson BAG, Rylander L, Lindh C, Rignell-Hydbom A, Giwercman A, Toft G et al (2005) Inter-population variations in

123

concentrations, determinants of and correlations between 2,2’,4,4’,5,5’-hexachlorobiphenyl (CB-153) and 1,1-dichloro2,2-bis (p-chlorophenyl)-ethlyene (p, p’-DDE): a cross sectional study of 3161 men and women from Inuit and European populations. Environ Health 4:27 Koppen G, Covaci A, Van Cleuvenbergen R, Schepens P, Winneke G, Nelen V et al (2002) Persistent organochlorine pollutants in human serum of 50-65 years old women in the Flanders Environmental and Health Study (FLEHS). Part 1: concentrations and regional differences. Chemosphere 48:811–825 Kvalem HE, Knutsen HK, Thomsen C, Haugen M, Stigum H, Brantsaeter AL et al (2009) Role of dietary patterns for dioxin and PCB exposure. Mol Nutr Food Res 53:1438–1451 LaKind JS, Berlin CM, Naiman DQ (2001) Infant exposure to chemicals in breast milk in the Unites States: what we need to learn from a breast milk monitoring program. Environ Health Perspect 109:75–88 Lino CM, da Silveira MI (2006) Evaluation of organochlorine pesticides in serum from students in Coimbra Portugal: 1997-2001. Environ Res 102:339–351 Llop S, Ballester F, Vizcaino E, Murcia M, Lopez-Espinosa MJ, Rebagliato M et al (2010) Concentrations and determinants of organochlorine levels among pregnant women in Eastern Spain. Sci Total Environ 408:5758–5767 Luzardo OP, Goethals M, Zumbado M, Alvarez-Leo´n EE, Cabrera F, Serra-Majem L et al (2006) Increasing serum levels of nonDDT-derivates organochlorine pesticides in the younger population of the Canary Islands (Spain). Sci Total Environ 67:129–138 Moysich KB, Ambrosone CB, Mendola P, Kostyniak PJ, Greizerstein HB, Vena JE et al (2002) Exposures associated with serum organochlorine levels among postmenopausal women from western New York State. Am J Ind Med 41:102–110 Nøst TH, Breivik K, Fuskeva˚g OM, Nieboer E, Odland JØ, Sandanger TM (2013) Persistent organic pollutants in Norwegian men from 1979 to 2007: intraindividual changes age-period-cohort effects and model predictions. Environ Health Perspect 121:1292–1298 Park H, Lee SJ, Kang JH, Chang YS (2007) Congener-specific approach to human PCB concentrations by serum analysis. Chemosphere 68:1699–1706 Patterson DG, Isaacs SG, Alexander LR, Turner WE, Hampton L, Bernert JT et al (1991) Determination of specific polychlorinated dibenzo-p-dioxins and dibenzofurans in blood and adipose tissue by isotope dilution-high-resolution mass spectrometry. IARC Sci Publ 108:299–342 Petrik J, Drobna B, Pavuk M, Jursa S, Wimmerova S, Chovancova J (2006) Serum PCBs and organochlorine pesticides in Slovakia: age gender and residence as determinants of organochlorine concentrations. Chemosphere 65:410–418 Philips DL, Pirkle JL, Burse VW, Bernert JT Jr, Henderson LO, Needham LL (1989) Chlorinated hydrocarbon levels in human serum: effects of fasting and feeding. Arch Environ Contam Toxicol 18:495–500 Pinheiro JC, Bates DM (2000) Mixed-effects models in S and S-PLUS. Springer, New York Porta M, Puigdome`nech E, Ballester F, Selva J, Ribas-Fito´ N, Llop S et al (2008) Monitoring concentrations of persistent organic pollutants in the general population: the international experience. Environ Int 34:546–561 Porta M, Gasull M, Puigdome`nech E, Garı´ M, Bosch de Basea M, Guille´n M et al (2010) Distribution of blood concentrations of persistent organic pollutants in a representative sample of the population of Catalonia. Environ Int 36:655–664 Porta M, Lo´pez T, Gasull M, Rodrı´guez-Sanz M, Garı´ M, Pumarega J et al (2012) Distribution of blood concentrations of persistent organic pollutants in a representative sample of the population of

Arch Environ Contam Toxicol Barcelona in 2006 and comparison with levels in 2002. Sci Total Environ 423:151–161 Rogan WJ, Chen A (2005) Health risks and benefits of bis(4chlorophenyl)-1,1,1-trichloroethane (DDT). Lancet 366:763–773 Rylander C, Sandanger TM, Brustad M (2009) Associations between marine food consumption and plasma concentrations of POPs in a Norwegian coastal population. J Environ Monit 11:370–376 Schisterman EF, Whitcomb BW, Louis GM, Louis TA (2005) Lipid adjustment in the analysis of environmental contaminants and human health risks. Environ Health Perspect 113:853–857 Sjo¨din A, Jones RS, Focant JF, Lapeza C, Wang RY, McGahee EE 3rd et al (2004) Retrospective time-trend study of polybrominated diphenyl ether and polybrominated and polychlorinated biphenyl levels in human serum from the United States. Environ Health Perspect 112:654–658 Tee PG, Sweeney AM, Symanski E, Gardiner JC, Gasior DM, Schantz SL (2003) A longitudinal examination of factors related to changes in serum polychlorinated biphenyl levels. Environ Health Perspect 111:702–707 Vizcaino E, Grimalt JO, Carrizo D, Lopez-Espinosa MJ, Llop S, Rebagliato M et al (2011) Assessment of prenatal exposure to persistent organohalogen compounds from cord blood serum analysis in two Mediterranean populations (Valencia and Menorca). J Environ Monit 13:422–432

Welch AA, Lund E, Amiano P, Dorronsoro M, Brustad M, Kumle M et al (2002) Variability of fish consumption within the 10 European countries participating in the European Investigation into Cancer and Nutrition (EPIC) study. Public Health Nutr 5:1273–1285 Wolff MS, Britton JA, Teitelbaum SL, Eng S, Deych E, Ireland K et al (2005) Improving organochlorine biomaker models for cancer research. Cancer Epidemiol Biomark Prev 14:2224–2236 Zubero MB, Aurrekoetxea JJ, Ibarluzea JM, Arenaza MJ, Rodrı´guez C, Sa´enz JR (2010) Heavy metal levels (Pb Dd Cr & Hg) in the adult general population near an urban solid waste incinerator. Sci Total Environ 408:4468–4474 Zubero MB, Aurrekoetxea JJ, Ibarluzea JM, Rivera J, Parera J et al (2011) Evolution of PCDD/Fs and dioxin-like PCBs in the general adult population living close to a MSW incinerator. Sci Total Environ 410:241–247 Zumbado M, Goethals M, Alvarez-Leo´n EE, Luzardo OP, Cabrera F, Serra-Majem L et al (2005) Inadvertent exposure to organochlorine pesticides DDT and derivatives in people from the Canary Islands (Spain). Sci Total Environ 339:49–62

123

Time trends in serum organochlorine pesticides and polychlorinated biphenyls in the general population of Biscay, Spain.

Despite the measures adopted, levels of organochlorine compounds (OCs) are still being detected in the human body. This study aimed to explore factors...
494KB Sizes 0 Downloads 6 Views